chapter 1
The Hadza and
Evolutionary Theory
An Introduction
The Hadza of Tanzania are one of the very few societies anywhere in the
world who still live by hunting and gathering. Hunter-gatherers are
people who forage for wild foods, practicing no cultivation or animal
husbandry. The fact that the Hadza are still foraging makes them invaluable to researchers interested in the lifestyle of our ancestors before agriculture so greatly altered human societies. The Hadza happen to live in
East Africa, an area rich in hominin fossils (Figure 1.1). Hominins are
all those species (Australopithecus afarensis, Paranthropus boisei, Homo
erectus, etc.) that share with us a common ancestor that diverged from
the ancestor of our closest living relatives, chimpanzees and bonobos, 6–7
million years ago (mya). Today, we humans are the sole surviving hominin.
Humans and their hominin ancestors have occupied East Africa for as
long as they have existed. For testing evolutionary hypotheses about past
ecological influences and outcomes, it would be difficult to find a living
society more relevant than the Hadza. They have been studied extensively,
but surprisingly no English-language anthropological book has been published until now.1
I began my research with the Hadza in 1995 when I was working on
my Ph.D. at UCLA. Since then, I have been back 15 times for a total of
about 4 years altogether. Even so, I speak only a little bit of the Hadza
1. In 1958 Kohl-Larsen wrote a book in German, but it has not been translated.
1
2
The Hadza and Evolutionary Theory
Hadar
Gulf of Aden
Bodo
N
SOMALIA
Gona
ETHIOPIA
Serengeti National Park
0 10 20 30 40 km
SUDAN
Ngorongoro Crater
Omo
MASAI
Koobi Fora
SUKUMA
Karatu
Endamagha
Oldeani
Lake
Turkana
Maswa Game Preserve
MANGOLA
Lake
Eyasi
DUNDUIYA
UGANDA
Tugen Hills
HADZA
KENYA
Chesowanja
Matala
IRAMB A
TLIíIKA
DATOGA
Mongo wa Mono
Mbulu
Yaeda Chini
Munguli
Olorgasailie
Lake
Victoria
Lake
Manyara
SIPUNGA
IRAQW
Peninj
ISANZU
Haidom
Olduvai Gorge
Laetoli
Mumba
HADZA
INDIAN OCEAN
0
100 200 300 km
TAN ZANIA
Figure 1.1. Map of East Africa, showing Hadza area in relation to the locations
of hominin fossil sites. Inset shows Hadzaland (inside shaded ring) with key
locations in a small font and names of neighboring ethnic groups in a large
font.
language (Hadzane) because I can communicate with the Hadza in
Kiswahili, their second language. My research entails behavioral sampling, measurement of food acquisition and consumption, skills, preferences, anthropometry, demography, and a variety of topics using experiments and interviews. All of this research involves evolutionary theory
as a guide to hypothesis testing. I intend this book to serve as a general
ethnography of the Hadza, but I also hope to persuade the uninitiated
that an evolutionary view is essential for understanding humans. I therefore aim to make the book accessible to readers with little familiarity
with evolutionary theory.
The Hadza and Evolutionary Theory
3
Natural selection explains how and why evolution occurs (Darwin
1859). Individuals with certain traits survive and reproduce more successfully than individuals with other traits. The genes involved in producing
those successful traits are passed on to the next generation in greater
numbers. Darwin did not know about genes, but he knew traits were inherited; the modern theory of selection had to await the rediscovery of
Mendelian genetics and the synthesis of Mendel’s laws with Darwin’s
theory by early population geneticists (Fisher 1930, Haldane 1932, Wright
1931). Natural selection leads to adaptation. Genes that promote more
optimal use of resources in an environment result in enhanced survival
and reproduction of their carriers, and consequently each successive generation becomes more adapted to the environment. When the environment
changes, the population will eventually track the changes or go extinct.
The environment includes everything: climate, flora and fauna, predators,
parasites, and members of one’s own species (conspecifics). Conspecifics
are especially important among social animals like us because kin and
potential rivals, allies, and mates have a strong impact on an individual’s
reproductive success (RS).
Our bodies and behavior have been shaped by competition between
individuals in the past, and we all have inherited our genes from the winners of that competition. Sexual selection is the component of natural
selection that deals with reproduction (Darwin 1871). Ultimately, greater
RS matters more than longevity because no one lives forever. Even if one
lived to be a thousand years old, without reproducing, one’s own genes
would not be passed on, so one’s personal (or direct) fitness would be zero.
We can see, therefore, why selection should commonly favor individuals
who selfishly exploit resources for their own reproductive benefit (individual selection).
Our understanding of how evolution works greatly expanded when
William Hamilton (1964) explained why, despite individual selection, we
observe seemingly altruistic behavior in many species. Hamilton’s formula
tells us when selection should favor one individual helping another. The
formula can be expressed as C < Br, where C equals the cost to the helper,
B equals the benefit to the beneficiary, and r equals the degree of relatedness between helper and beneficiary. When the helper and beneficiary are
related by 50%, as full siblings are on average, as long as the benefit is
more than twice the cost (e.g., benefit = 2.1 and cost = 1), it will pay one
sibling to help the other because 0.5* 2.1 = 1.05 > 1. More precisely, there
is positive selection for a gene that promotes helping relatives (nepotism).
Hamilton (1964) thus introduced the concept of inclusive fitness. An
4
The Hadza and Evolutionary Theory
individual’s inclusive fitness includes the person’s own RS (direct fitness)
plus his or her effect on others who carry the same alleles by common descent, weighted by the degree of relatedness between them (indirect fitness)
(Creel 1991, Hamilton 1964, Lucas, Creel, and Waser 1996).
The reason Hamilton’s paper was such a profound development was
that it explained seemingly altruistic behavior without invoking group
selection. One proponent of group selection, Wynne-Edwards (1962)
had suggested that a population of birds restrains its reproduction to
avoid overexploitation of resources and to prevent group extinction. But
David Lack’s (1966) experiments had already shown that optimal clutch
size, rather than maximum clutch size, was favored because a female
who laid too many eggs fledged fewer chicks than when she laid the optimal number. G. C. Williams (1966) made the case that Lack’s explanation
of reproductive restraint was more consistent with basic Darwinian theory
because restraint increased the individual mother’s RS.
Darwin’s natural selection was fundamentally about individuals competing for resources and turning them into more offspring than their rivals had (individual selection). Nonetheless, before Hamilton’s seminal
paper, one commonly heard group selection arguments like “subordinates
defer to dominant individuals” or “the slowest animal gets eaten by the
predator” because it is “for the good of the species.” Even Darwin himself invoked group selection to explain behavior seen in warfare, saying
that those who were bravest, risking death to save their group, would
make the best warriors and that self-sacrificing behavior would evolve
from their defeating other groups (Darwin 1871). On the other hand, he
had no explanation for the existence of sterile castes within bees, wasps,
and ants, but recognized that their inability to reproduce posed a potentially fatal blow to his theory of natural selection that depended on reproduction. Hamilton’s formula provided an explanation: The sterile
workers helped the queen make more sisters for them because given their
haplo-diploid genetic system they were more closely related to their sisters
(r = 75%). This explanation highlighted that it is the gene, not the group
or the individual, that is selected. Because genes behave as if they are selfish,
individuals sometimes behave unselfishly (Dawkins 1976).
No one ever puzzled over why a mother was willing to risk her life to
protect her offspring; they are her genes’ vehicles into the future. Helping other kinds of kin can be favored for the same reason, even though
the path to the next generation is less direct. Hamilton’s rule tells us that
even a mother will sometimes fail to help her offspring. Her interests and
those of her offspring do not overlap perfectly because she shares only
The Hadza and Evolutionary Theory
5
50% of her genes with each one. She should weigh the gains of helping
one offspring against the costs to her future offspring. In other words,
we can expect parent-offspring conflict (Trivers 1974). A mother shares
50% of her genes with each of her two children, so she wants to help
both equally. On the other hand, the child shares 50% of its genes with
its full sib, which is only half of the 100% in its own genome. The child
will prefer to get more help from mother than is in the mother’s best interest or the interest of the other sib. The ages of individuals will also affect the calculus because a very young sib may not be capable of helping
its older sib, while the elder may be able to help the younger.
It is inclusive fitness that determines a gene’s success in succeeding
populations, more precisely, the success of particular variants of the genes
called alleles. There may be a brown or blue allele occupying the site of
the gene coding for eye color. Often one allele is so successful that it
spreads throughout the population and replaces all alternatives, at which
point it is “fixed.” By influencing the individual’s behavior in ways that
cause them to get passed on, genes affect their own representation in the
gene pool. From the perspective of the allele, any behavior that helps
make copies in others is advantageous, even if it causes its host to behave
in an unselfish, even self-destructive way, so long as this leaves the greatest number of copies in the next generation. However, since genes in the
individual’s genome share their fate and will all die together, they tend
to work well together; otherwise they are less likely to be passed on. For
this reason, genes tend to promote a considerable degree of self-regarding
behavior in individuals. Dawkins (1976) used the metaphor of rowers in
a boat; they go nowhere if they do not paddle in concert and in the same
direction.
How animals meet their energetic and nutritional requirements is the
focus of optimal foraging theory (OFT). OFT assumes that individuals
will seek to minimize costs and maximize benefits in their foraging behavior. That means they will usually not waste time or energy going
after low-yield foods when higher-yield foods are available. They should
be efficient in harvesting energy, and energy is used as a proxy for fitness. However, the individual always faces many constraints. The first
constraint is the diet itself; committed carnivores cannot live on plant
foods and herbivores on meat. Other important questions include the
proper currency to use (kilocalories vs. protein) and the primary goal
(maximize average rate of energy intake vs. minimize risk of shortfall)
(Stephens and Krebs 1986). OFT models, like models of morphological
evolution, are essentially engineering problems. With a given availability
6
The Hadza and Evolutionary Theory
table 1.1. payoff matrix for
hawk- dove game
H
D
H
(prob. win * value of food) +
(prob. lose * cost of injury)
(0.5 * 50) + (0.5 * −100) =
25 − 50 = −25
(prob. win * value of food)
(1 * 50) = 50
D
(prob. win * value of food)
(0 * 50) = 0
(prob. win * value of food) +
(prob. lose * cost of display)
(0.5) * (50 − 10) +
(0.5) * (−10) = 20 − 5 = 15
note: Payoffs shown are for the player listed in the row.
of food A and food B, each with a different nutritional value and a required search and handling time, how much of A vs. B should a forager
take? But in life there are always competitors, and we need to consider
their effect. Game theory is ideally suited for this.
Game theory treats individuals as different strategies, lets them play
against each other in a game with specified payoffs for certain outcomes,
and then sees which strategy wins. In the hawk-dove game, two strategies are employed with regard to a food source. A hawk is willing to
fight and risk injury to get the food. A dove only displays, which goes on
a while against another dove, but gives up quickly against a hawk. When
dove meets dove, each gets the food half the time. If the value of the food
is worth a gain of 50 points, the cost of a dove’s long display is a loss of
10 points, and the cost of injury in a fight is a loss of 100 points, we can
calculate the outcome (Table 1.1). When doves meet, each has a 50-50
chance of getting the food, so each receives an average of 25 minus the
cost of display (25 – 10 = 15). When a hawk meets a dove, hawk gets the
food and 50 points. When two hawks meet, they each have the same chance
of winning their fight and getting the food but also the same chance of
being injured. That gives each an average of 25 for getting the food and
–50 for getting injured; thus their average payoff is –25.
In each instance, the hawk beats the dove, which means hawks can always invade a population of doves and do very well (+50). However,
when there are too many hawks, they do poorly (–25). Doves do better
against each other (+15); even against a hawk, doves get 0, which is better than two hawks against each other (–25). This means that no pure
strategy is stable. With the payoffs specified here, there will be a mixed
The Hadza and Evolutionary Theory
7
strategy with an equilibrium frequency of 0.58 hawk and 0.42 dove.
This might consist of a population of 58% hawks and 42% doves, or individuals who play the hawk strategy 58% of the time and the dove
strategy 42% of the time. The strategy that does best against other strategies and against itself, and therefore cannot be invaded by other strategies,
is called an evolutionarily stable strategy (ESS) (Maynard Smith 1982). A
strategy called “bourgeois,” in which one defends food if there first or if
closest to the food, is an ESS because it avoids fights against other bourgeois by using a simple, salient rule—proximity or priority—yet does not
cause one to relinquish food like a dove does against a hawk (Maynard
Smith 1982).
Genetic selection leads to adaptive behavior across environments, but
because optimal solutions vary in different environments and different
circumstances (different games), so too does behavior. Although we do
not know precisely how alleles at genetic loci translate into specific behaviors, we can assume behavioral strategies, or decision rules (which
do not imply conscious decisions), have been favored by selection to
create adaptive phenotypes. This assumption of a link between heritability and phenotype, known as the phenotypic gambit (Grafen 1984,
Smith and Winterhalder 1992), allows us to make predictions about behavioral outcomes. That is, we expect to see individuals modify their
behavior in ways that tend to enhance their inclusive fitness. Just as our
bodies tell us to seek water to quench our thirst, so our emotions tell us to
flee an enemy, recruit allies, or seek a mate. Behavioral ecologists tend
to measure behavior directly to test predictions about fitness outcomes.
For this reason, they have mostly studied simpler societies (Borgerhoff
Mulder 1988a, Winterhalder and Smith 2000), not necessarily foragers,
but at least natural fertility (non-contracepting) societies, where there is
still a direct link between behavior and RS.
Evolutionary psychology focuses on the mind and the mechanisms
that selection has favored to execute adaptive behavior (Hagen 2003).
Such mechanisms are assumed to be universal across humans, by and
large. This means humans anywhere are appropriate subjects. Much depends on the trait, however. Some traits, like fear of snakes or desire for
sex, may have changed little over the last several million years, while
others, such as tribal identification and loyalty or the value placed on
fatness in potential partners, may have changed significantly even over
the past few hundred years.
The Environment of Evolutionary Adaptedness (EEA) (Bowlby 1969)
refers to the sum total of environments hominins have occupied since
8
The Hadza and Evolutionary Theory
they diverged from the ancestor of chimpanzees and bonobos 6–7 mya.
It is within this time period that all hominin traits that were not present
in the last common ancestor must have evolved. These are called derived
traits; ancestral traits are those that were present in the last common
ancestor. The EEA obviously includes a wide range of different environments, and different ones are relevant for different traits. The relevant
environment for the evolution of our lungs dates back to the Devonian
about 400 mya, for bipedalism to the end of the Miocene 6–7 mya,
and for our peculiarly enlarged brain to about 2.5 mya. Lactose tolerance (the ability to digest milk in adults) dates to only about 7,000 years
ago, and only in certain populations (Bersaglieri et al. 2004 ). Given that
the relevant period varies with the trait in question, it makes more sense
to think in terms of the Adaptively Relevant Environment (ARE) (Irons
1998).
Culture is obviously an important component of the human environment. When we talk about different environments, we are not simply
contrasting desert with rain forest, but forager camp with inner-city neighborhood, societies where marriage is arranged by parents with those where
people pick their own mates, and societies where food is scarce with societies in which food is guaranteed by a welfare program. Humans have
created such varied cultural environments that it is often culture, rather
than the physical habitat, that determines which behaviors enhance fitness. Ignoring culture can leave us without a clue (Laland, Odling-Smee,
and Feldman 2000). However, there is something unsatisfying about an
explanation of one cultural trait by reference to another without any attempt to explain how the first trait came about. For example, people often attribute sexual chastity in some cultures to adherence to religious
doctrine. Religion may well be a proximate explanation, but why did a
strict religious doctrine evolve in one population and not in another?
Here, I attempt to see how much we can understand by reference to the
habitat and mode of subsistence. Since among forgers each person must
spend much time looking for and processing foods, it is easier to link behavior to the habitat and ecology.
The novel environments we have created can give rise to maladaptive
behavior. Maladaptive does not mean bad, only fitness-reducing. This can
occur when there has not been sufficient time for selection to produce the
appropriate cognitive processing of the new environmental cues. Thus, individuals cannot calibrate the appropriate behavioral response that would
maximize inclusive fitness (Richerson and Boyd 2005). It is important to
appreciate that decision rules need not be conscious to work, and often it
The Hadza and Evolutionary Theory
9
would be inefficient if they were. Imagine a truck coming at you; it is best
if you do not have to consciously deliberate and decide whether to jump
out of the street. When past selection produced adaptive behavior that is
no longer fitness-enhancing in the present, it is called a mismatch. To cite
one example, a study found that wealthier men in Montreal did not have
more children, though they did have more sexual partners (Perusse 1994).
Before contraception, more sexual partners would likely have resulted in
more children. Note that what was selected for in the past was desire for
sexual partners, not desire for siring offspring. Still, selection never ceases.
As long as some people do want to have some children someday (and most
people do), reproductive competition continues.
Mismatch is less of an issue among foragers because they are still living in an environment without so many novel cues, and we can expect
mostly adaptive behavior from them (Symons 1987, Tooby and Cosmides 1990). There is still much to learn from foragers like the Hadza.
Because they live in an area our hominin ancestors inhabited for as long
as they existed, where the flora and fauna have not changed much since
the Pliocene began 5.3 mya, their ecology is relevant for a long period of
human evolution. Mortality risks, energy expenditure, and growth, as
we will see in Chapter 6, are probably not so different from what they
were 10 –20,000 or more years ago. Because the Hadza live outdoors,
most of what they do is in full view, which means their behavior can be
measured directly. The foods they acquire can be weighed and, with effort, traced to the person who consumes them. We can see what sort of
help a parent can give that might impact survivorship or growth. There
are few books that report on quantitative behavioral studies carried out
in foraging societies (Hewlett 1991, Hill and Hurtado 1996, Howell
1979, Lee 1984). We now have a record of the Hadza going back to the
late 1950s, with some continuity back to the 1930s and even a bit earlier. A book-length treatment of Hadza research is long overdue.
This page intentionally left blank